
* OPEN STATA OUTPUT FILE LOG *

   log using "C:\Users\gk57526\Dropbox\Informal & Formal Bargaining (BAA Reforms)\JOP Replication File Materials\Output\execbudinfluence.08-26-2020.SECTION 6.ALTERNATIVE EXPLANATIONS FOR AGENCY HEAD TENURE EFFECTS.comments.smcl" 


   **** PROGRAM FILE FOR "INFORMAL VERSUS FORMAL BARGAINING STRUCTURES" PROJECT:  EMPIRICAL ANALYSES MOTIVATED BY COMMENTS ON VERSION 1.0  ****
 
 
   **** NOTE: DROP EXECUTIVE BRANCH AGENCY (AGENID==15) FROM THE ANALYSIS SINCE IT SERVES THE PRESIDENT, AND HENCE AHT IS SIMPLY THE PRESIDENT'S TENURE IN OFFICE ****


   
   
 * OPEN "EXECUTIVE BUDGETARY INFLUENCE" DATABASE WITH NEWLY CREATED VARIABLES
 
use "C:\Users\gk57526\Dropbox\Informal & Formal Bargaining (BAA Reforms)\JOP Replication File Materials\Data\execbudinfluence.08-26-2020.REPORTED MODELS.REV.comments.dta", replace

 

*** DECLARE DATABASE AS A PANEL (AGENCY ID * FISCAL YEAR): 

xtset agenid fyear, yearly

*
*
*



*** SECTION 5. RULING OUT ALTERNATIVE EXPLANATIONS FOR AGENCY HEAD TENURE EFFECTS [PROPOSITION # 3]: 



************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************



*** SYNC TO COMPARABLE ESTIMATION SAMPLE USED IN ENDOGENOUS REGIME SWITCHING REGRESSION MODELS USED TO TEST HYPOTHESES IN MANUSCRIPT ***


quietly movestay (congbudchangereg2 l1congbudchangereg2  execbudchangereg2 bureauofbudget ebcreg2baa lnagencyheadtenure  baalnaht  ebcreg2lnaht  ebcreg2baalnaht  ///
lnagencyage unifiedpartygovt presadminyear) if agenid!=15, cluster(agenid) select(regimereg = executiveagency) difficult


**********************************************************************************************************************************************************************************************************
**********************************************************************************************************************************************************************************************************
**********************************************************************************************************************************************************************************************************
**********************************************************************************************************************************************************************************************************




****** 6A.  EVALUATE WHETHER AGENCY HEAD TENURE IS LONGER DURING PRE-BAA REFORM ERA THAN COMPARED TO POST-BAA REFORM ERA? ***


***  WHY?  Influence of Agency Head Tenure May Partially Reflect Longer Tenures in Pre-Reform Era than Post-Reform Era [i.e., 'lack balance'] ///
***        Alternative Explanation to Reported Findings Rooted in Differential Tenure Effects Between Eras   



*** TABLE SI-6: FULL SAMPLE PERIOD (FY 1894 - FY 1940)  [SECTION 6: SUPPORTING INFORMATION DOCUMENT] -- IS AHT LONGER IN PRE-REFORM ERA THAN POST-REFORM ERA? 

ttest agencyheadtenure if e(sample), by(bureauofbudget) unequal

*** RESULT: AHT IS MARGINALLY [p = 0.0781] GREATER IN POST-REFORM ERA COMPARED TO PRE-REFORM ERA -- RUNS COUNTER TO ALTERNATIVE EXPLANATION NOTED ABOVE ***


********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************



*** TABLE SI-6: RESTRICTED SAMPLE PERIOD # 1 (FY 1910 - FY 1940)  [SECTION 6: SUPPORTING INFORMATION DOCUMENT] --- ///
***            IS AHT LONGER IN PRE-REFORM ERA THAN POST-REFORM ERA [RESTRICTED SAMPLE PERIOD # 1 WINDOW: FY 1910 - FY 1940]? 

ttest agencyheadtenure if e(sample) & fyear>=1910 & fyear<=1940, by(bureauofbudget) unequal

*** RESULT: AHT IS MARGINALLY [p = 0.0993] GREATER IN POST-REFORM ERA COMPARED TO PRE-REFORM ERA -- RUNS COUNTER TO ALTERNATIVE EXPLANATION NOTED ABOVE ***



********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************


*** TABLE SI-6: RESTRICTED SAMPLE PERIOD # 2 (FY 1910 - FY 1930)  [SECTION 6: SUPPORTING INFORMATION DOCUMENT] --- ///
***             IS AHT LONGER IN PRE-REFORM ERA THAN POST-REFORM ERA [RESTRICTED SAMPLE PERIOD # 1 WINDOW: FY 1910 - FY 1930]? 

ttest agencyheadtenure if e(sample) & fyear>=1910 & fyear<=1930, by(bureauofbudget) unequal

*** RESULT: AHT IS NOT MARGINALLY [p = 0.185] GREATER IN PRE-REFORM ERA COMPARED TO POST-REFORM ERA -- NULL DIFFERENCE ***



********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************


*** TABLE SI-6: RESTRICTED SAMPLE PERIOD # 3 (FY 1919 - FY 1927)  [SECTION 6: SUPPORTING INFORMATION DOCUMENT] --- /// 
***              IS AHT LONGER IN PRE-REFORM ERA THAN POST-REFORM ERA [RESTRICTED SAMPLE PERIOD # 1 WINDOW: FY 1919 - FY 1927]? 

ttest agencyheadtenure if e(sample) & fyear>=1919 & fyear<=1927, by(bureauofbudget) unequal

*** RESULT: AHT IS NOT MARGINALLY [p = 0.288] GREATER IN POST-REFORM ERA COMPARED TO PRE-REFORM ERA -- NULL DIFFERENCE  ***



********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************






****** 6B.  EVALUATE WHETHER AGENCY HEAD TENURE IS LONGER FOR INDEPENDENT AGENCIES COMPARED TO EXECUTIVE DEPARTMENTS DURING PRE-BAA REFORM ERA /// 
******      [AND VICE-VERSA DURING POST-REFORM ERA]?  ***

***       Is AHT longer for IAs Pre-Reforms, and longer for EDs Post-Reform? 
***       WHY?  Address Whether Reported Findings Are Due to Agency Design Differences, and not AHT, is the mechanism at work explaining EBI. ***  




*** 'SECOND' POINT TEXT DISCUSSION ON PAGE 21, IMMEDIATELY FOLLOWING TABLE SI-6 [SECTION 6: SUPPORTING INFORMATION DOCUMENT] ***

ttest agencyheadtenure if e(sample) & bureauofbudget==0, by(executiveagency) unequal



*** 'SECOND' POINT TEXT DISCUSSION ON PAGE 21, IMMEDIATELY FOLLOWING TABLE SI-6 [SECTION 6: SUPPORTING INFORMATION DOCUMENT] **

ttest agencyheadtenure if e(sample) & bureauofbudget==1, by(executiveagency) unequal



*** RESULTS:  AHT IS SIGNIFICANTLY SHORTER FOR EXECUTIVE DEPARTMENTS COMPARED TO INDEPENDENT AGENCIES FOR POST-REFORM ERA [CONTRARY TO CONCERNS NOTED ABOVE] ***
***           AHT IS NOT SIGNIFICANTLY DIFFERENT FOR EXECUTIVE DEPARTMENTS COMPARED TO INDEPENDENT AGENCIES DURING PRE-REFORM ERA [CONTRARY TO CONCERNS NOTED ABOVE]   



********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************





*** BIVARIATE RELATIONSHIP BETWEEN AGENCY HEAD TENURE & AGENCY SIZE BOTH IN NATURAL LOGARITHM FORMS] ****

generate lnrealcongappreg = ln(realcongappreg)
*
generate lnrealcongapptot = ln(realcongapptot)

*
*
*

*** FULL SAMPLE PERIOD: 'THIRD/FINALLY' POINT TEXT DISCUSSION ON BOTTOM PORTION OF PAGE 21 [SECTION 6: SUPPORTING INFORMATION DOCUMENT] **

regress lnagencyheadtenure lnrealcongappreg if e(sample) 
*
regress lnagencyheadtenure lnrealcongapptot if e(sample) 
*
*
*

*** PRE-BAA REFORM PERIOD: 'THIRD/FINALLY' POINT TEXT DISCUSSION ON BOTTOM PORTION OF PAGE 21 [SECTION 6: SUPPORTING INFORMATION DOCUMENT] **

regress lnagencyheadtenure lnrealcongappreg if e(sample) & bureauofbudget==0
*
regress lnagencyheadtenure lnrealcongapptot if e(sample) & bureauofbudget==0
*
*
*

*** POST-BAA REFORM PERIOD: 'THIRD/FINALLY' POINT TEXT DISCUSSION ON BOTTOM PORTION OF PAGE 21 [SECTION 6: SUPPORTING INFORMATION DOCUMENT] **

quietly movestay (congbudchangereg2 l1congbudchangereg2  execbudchangereg2 bureauofbudget ebcreg2baa lnagencyheadtenure  baalnaht  ebcreg2lnaht  ebcreg2baalnaht  ///
lnagencyage unifiedpartygovt presadminyear) if agenid!=15, cluster(agenid) select(regimereg = executiveagency) difficult
*
*
regress lnagencyheadtenure lnrealcongappreg if e(sample) & bureauofbudget==1
*
regress lnagencyheadtenure lnrealcongapptot if e(sample) & bureauofbudget==1








log close
